RecSysMisInfo

Data and source code of the experiments reported in our work published in WebSci '24. If you use our source code, dataset, or experiments for your research or development, please cite the following paper:

@inproceedings{DBLP:conf/websci/FernandezBC24,
  author       = {Miriam Fern{\'{a}}ndez and
                  Alejandro Bellog{\'{\i}}n and
                  Iv{\'{a}}n Cantador},
  title        = {Analysing the Effect of Recommendation Algorithms on the Spread of
                  Misinformation},
  booktitle    = {Proceedings of the 16th {ACM} Web Science Conference, {WEBSCI} 2024,
                  Stuttgart, Germany, May 21-24, 2024},
  pages        = {159--169},
  publisher    = {{ACM}},
  year         = {2024},
  url          = {https://doi.org/10.1145/3614419.3644003},
  doi          = {10.1145/3614419.3644003}
}

Code

The source code used in these experiments was written in Java and heavily depends on the RankSys library.

The starting point is the script file run_exp.sh, where all the necessary steps to reproduce the results presented in the paper are included. The actual Java code is in the src folder, and the packaged JAR file can be found inside the target folder.

Data

The data used in this work, as explained in the paper, includes user interactions from X (previously Twitter) and misinforming claims. The claims were collected by merging information from several datasets:

• CoronaVirusFacts Alliance
• Misinfo.me
• Covid-19 two myths
• CMU-MisCov19: paper and data

The generated dataset can be found in the generated_dataset folder. While the actual files used by the recommendation algorithms can be found in the root folder of the repository, with names starting with 'merged__user_item'.

Authors

• Miriam Fernández, The Open University, United Kingdom
• Alejandro Bellogín, Universidad Autónoma de Madrid, Spain
• Iván Cantador, Universidad Autónoma de Madrid, Spain